Download Intelligent Systems for Animal Welfare

Charting Unconquered Territories:
Intelligent Systems for Animal Welfare
Clara Mancini1 and Anna Zamansky 2
Abstract. In introducing the First Symposium of Intelligent Systems for Animal Welfare, we share our vision for a new multidisciplinary field of animal welfare informatics, in which scholars from
artificial intelligence, computer science, animal science, veterinary
medicine and other fields join efforts to exploit state-of-the-art technologies for the improvement of animal welfare.
1
Animal Welfare
The digital era has revolutionized our quality of life in every possible
aspect. Intelligent systems which help us make better decisions, analyze big data and manage complex information are omnipresent in
healthcare, education, commerce and many other fields. But are the
recent developments realizing their full potential in a similar way in
the lives of animals? Can we take advantage of state-of-the-art scientific achievements in informatics and artificial intelligence to improve animal welfare? What constitutes good welfare for animals is
the object of ongoing research within animal welfare science (e.g.,
[10]). Generally, basic health and physiological functions, affective
states and the ability to express adapted behaviour are considered
important parameters for welfare assessment (e.g., [6]). In terms of
welfare standards, these dimensions are expressed by the following
principles of the Five Freedoms ([9, 1]):
1. Freedom from hunger or thirst by ready access to fresh water and
a diet to maintain full health and vigour;
2. Freedom from discomfort by providing an appropriate environment including shelter and a comfortable resting area;
3. Freedom from pain, injury or disease by prevention or rapid diagnosis and treatment;
4. Freedom to express (most) normal behaviour by providing sufficient space, proper facilities and company of the animal’s own
kind;
5. Freedom from fear and distress by ensuring conditions and treatment which avoid mental suffering.
Originated from a UK Government report on livestock husbandry
and subsequently developed by the British Farm Animal Welfare Advisory Committee, the Five Freedoms have since then been the base
of animal welfare standards defined by economic, scientific and advocacy organisations as diverse as the World Organisation for Animal Health, the American Veterinary Medical Association, the Royal
Society for the Prevention of Cruelty to Animals, to name just a few.
The above principles pose specific challenges for human practices
and their relations with animals. In particular, new scientific discoveries increasingly highlight the complexity of animals physiological
1
2
Open University, UK.
University of Haifa, Israel.
and psychological characteristics and welfare requirements. At the
same time, the issue of animal welfare is of growing concern to many
areas of society. Therefore there is a need for researchers to intensify
efforts aimed at improving the welfare of other beings in our dealings
with them.
2
Introducing ISAWEL’14
In initiating ISAWEL, a symposium on Intelligent Systems for Animal Welfare, our aim was to bring together animal scientists, who
could provide expertise on welfare problems, and computer scientists, who could provide insights into their possible technological
solutions. Our vision was to establish an ongoing interdisciplinary
dialogue to advance both animal welfare science on a computational
trajectory and the science of intelligent technology beyond humans.
As a starting point, this first edition of the symposium, ISAWEL14,
brings together research areas of Animal-Computer Interaction, neural networks and robotics through papers which are authored both by
animal experts and computer scientists.
The emerging area of Animal-Computer Interaction (ACI) [8] aims
to develop a user-centred approach to the design of technology intended for animals. ACI’s mission is to expand the boundaries of
interaction design and the design of interactive systems beyond the
human species. Its aims include investigating the interaction between
animals and technology, designing technology to support animals in
different contexts, and developing user-centred approaches to the design of technology intended for more-than-human animals [7]. ACI
has the potential to improve animal welfare in a variety of ways: for
example, by affording them more control over their own environment
(see ISAWEL14 paper by Mancini et al.) thus enabling the fulfilment
of their physiological needs as they arise; by providing them with
cognitive enrichment (e.g. see ISAWEL14 paper by French et al.)
thus promoting the expression of more natural behavior; or by supporting them in tasks they are employed to do (e.g. see ISAWEL14
paper by Robinson et al.) thus alleviating stress and distress.
As opposed to ACI, the field of artificial neural networks (ANNs)
has been present in AI for several decades. ANNs are computational
models inspired by animals’ central nervous systems (in particular
the brain) that are capable of machine learning and pattern recognition. In ISAWEL’14 paper by Barbuti et al., ANNs have been applied
to assist the population growth of tortoises by an automatic identification of nesting phases.
Finally, in a paper by Gergely et al., artificial agents are applied
for gaining insights into canine interaction with humans. Its authors
are leading experts from Family Dog Project ([2]), the first research
group dedicated to investigate the evolutionary and ethological foundations of dog-human relationship.
Framing the symposium’s five paper presentations, both keynotes
given by distinguished scholars in computer science and animal behavior demonstrate or directly argue for the need to take a multidisciplinary approach in animal welfare research.
The opening keynote is by David Harel, William Sussman Professor of Mathematics at the Weizmann Institute of Science, a recipient
of numerous prizes and distincions, including the prestigious Israel
and Emet Prizes. He has been working on a variety of topics within
computer science, including modeling and analysis of biological systems. His keynote demonstrates the fruitfulness of computational approaches when studying and modelling complex systems, such as the
behavior of animal populations, which is critical for understanding
their welfare requirements.
The closing keynote is given by Shaun Lawson and Daniel Mills.
Daniel Mills is Professor of Veterinary Behavioural Medicine and
Head of Research Strategy at the Univeristy’s of Lincoln’s School
of Life Sciences; he is a world-renouned expert in animal behaviour,
and has made key contributions to welfare practices for, among others, horses, dogs and cats. Shaun Lawson is Professor of Social Computing and Director of Lincoln Social Computing Research Centre (LiSC) and Director of Research at the University of Lincoln’s
School of Computer Science; he has collaborated with Daniel Mills
on numerous projects related to human-animal interactions. Their
keynote highlights the importance of collaborating with animal welfare scientists when developing technological interventions intended
for animal use.
To provide an opportunity for all symposium participants to take part
in the multidisciplinary dialogue to which presentations and keynotes
variously contribute, the programme includes a round table discussion on the themes emerged during the day.
3
Summary
As shown by the wide range of topics presented in these proceedings, multidisciplinary applications in animal welfare of methodologies and techniques from artificial intelligence, animal-computer interaction, robotics and intelligent systems are starting to emerge. We
believe that the time is ripe to start building a new discourse between
scholars from across a range of areas related to intelligent technologies and a range of areas pertaining to animal welfare to lay the foundations for a new discipline of animal welfare informatics.
4
Acknowledgements
ISAWEL’14 was supported by Universities Federation for Animal
Welfare (UFAW [3]) and Metabolic Robots ([4]). The latter is an Israeli start-up company, founded by Trendlines ([5]) in 2013, which
promotes animal welfare by developing a cloud-based automated
poultry feeding system which has been shown to reduce animal stress
levels during feedings and decrease mortality rates.
REFERENCES
[1]
[2]
[3]
[4]
[5]
[6]
http://www.defra.gov.uk/fawc/about/five-freedoms/.
http://familydogproject.elte.hu.
http://www.ufaw.org.uk.
http://www.metabolicrobots.com.
http://www.trendlines.com.
Vonne Lund, ‘Natural livinga precondition for animal welfare in organic farming’, Livestock Science, 100(2), 71–83, (2006).
[7] Clara Mancini, ‘Animal-computer interaction: a manifesto’, interactions, 18(4), 69–73, (2011).
[8] Clara Mancini, ‘Animal-computer interaction (aci): changing perspective on hci, participation and sustainability’, in CHI’13 Extended Abstracts on Human Factors in Computing Systems, pp. 2227–2236.
ACM, (2013).
[9] AJF Webster, ‘Farm animal welfare: the five freedoms and the free market’, The veterinary journal, 161(3), 229–237, (2001).
[10] Terry Whiting, ‘Understanding animal welfare: The science in its cultural context’, The Canadian Veterinary Journal, 52(6), 662, (2011).